Lazy release consistency for software distributed shared memory
ISCA '92 Proceedings of the 19th annual international symposium on Computer architecture
Simulation of multiprocessors: accuracy and performance
Simulation of multiprocessors: accuracy and performance
Virtual memory mapped network interface for the SHRIMP multicomputer
ISCA '94 Proceedings of the 21st annual international symposium on Computer architecture
Designing memory consistency models for shared-memory multiprocessors
Designing memory consistency models for shared-memory multiprocessors
The SPLASH-2 programs: characterization and methodological considerations
ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
Scope consistency: a bridge between release consistency and entry consistency
Proceedings of the eighth annual ACM symposium on Parallel algorithms and architectures
OSDI '96 Proceedings of the second USENIX symposium on Operating systems design and implementation
ISCA '99 Proceedings of the 26th annual international symposium on Computer architecture
Splitting the Data Cache: A Survey
IEEE Concurrency
Making Distributed Shared Memory Simple, Yet Efficient
HIPS '98 Proceedings of the High-Level Parallel Programming Models and Supportive Environments
Improving Release-Consistent Shared Virtual Memory using Automatic Update
HPCA '96 Proceedings of the 2nd IEEE Symposium on High-Performance Computer Architecture
| Hi-index | 0.00 |
Shared Virtual Memory (SVM) systems, because of their software implementation, enable shared memory programming at a low design and maintenance cost. Nevertheless, as hardware implementations become faster, their performance is still far from that achieved by distributed shared memory (DSM) systems. Nowadays, SVM use relaxed memory consistency models and multiple writer protocols as techniques to reduce latencies and false sharing, respectively; however, these techniques induce additional overhead that decreases system performance. In this paper, we perform a study of workload behavior aimed at improving the design of SVM protocols. The work focuses on the identification of the type of shared data patterns that can appear in the accesses to protected sections using semaphores. Most coherence actions in SVM systems are performed as a consequence of the write operations executed in critical sections, so this study pays special attention to the write operations performed when multiple writers are allowed. As these write operations may present spatial locality, we also study the write patterns on shared pages with similar behavior. Different software filters are applied in the instrumented parallel workloads selected to capture and classify the most common sharing patterns. This enables the recognition of those patterns in which coherence overhead can be reduced by modifying the coherence actions performed by the protocol. Despite the fact that performance evaluation of new coherence solutions is not the main goal of this paper, the ideas presented to improve the behavior of SVM systems can be implemented at a reasonable hardware/software cost.